Novel Quinolone Resistance Determinant, qepA8, in Shigella flexneri Isolated in the United States in 2016

LETTER

In Enterobacteriaceae, quinolone resistance is largely attributed to mutations in the quinolone resistance-determining regions (QRDRs) of gyrA, gyrB, parC, and parE and plasmid-mediated quinolone resistance (PMQR) genes [e.g., qnr genes, aac(6′)-Ib-cr, or qepA]. The functionality of four qepA genes, designated qepA1 to qepA4, has been described in the literature (1–5); however, other qepA genes were recently designated qepA5 to qepA10, despite having unknown functionality against quinolones (6). Here, we describe the genetic context and functionality of qepA8 detected in Shigella flexneri from a patient in the United States.

In 2016, routine genotypic and phenotypic characterization of Shigella isolates by the CDC National Antimicrobial Resistance Monitoring System (7) revealed qepA8 in a single ciprofloxacin-resistant Shigella flexneri strain (2016AM-0877; GenBank accession numbers CP033510.1 and CP033511.1). The strain was cultured from a stool sample from a patient who reported spending several months traveling in Taiwan, Thailand, Laos, Cambodia, and India. He had sought treatment in India for abdominal cramping/bloody diarrhea and was prescribed ciprofloxacin. His diarrhea worsened, and he sought medical care in the United States 2 to 3 weeks later. There, a Shigella sp. was isolated from his stool, and a 5-day course of ciprofloxacin was again prescribed. Phenotypic resistance was observed to amoxicillin-clavulanic acid, ampicillin, chloramphenicol, ciprofloxacin, nalidixic acid, streptomycin, and tetracycline (Table 1).

TABLE 1.

MICs and interpretations for antimicrobials tested against 2016AM-0877, E. coli TOP10, and transformant TOP10-pM216_mF

Antimicrobial	MIC (interpretation) (μg/ml) for straina:
	2016AM-0877	TOP10	TOP10-pM216_mF
Amoxicillin-clavulanic acid	32 (R)	NA	NA
Ampicillin	>32 (R)	NA	NA
Azithromycin	1 (WT)	NA	NA
Cefoxitin	8 (S)	NA	NA
Ceftriaxone	≤0.25 (S)	NA	NA
Chloramphenicol	>32 (R)	NA	NA
Ciprofloxacin	>4 (R)	≤0.0039 (S)	0.031 (S)
Levofloxacin	NA	0.0078 (S)	0.016 (S)
Gentamicin	1 (S)	NA	NA
Meropenem	≤0.06 (S)	NA	NA
Nalidixic acid	>32 (R)	NA	NA
Streptomycin	>64 (R)	NA	NA
Sulfisoxazole	≤16 (S)	NA	NA
Tetracycline	>32 (R)	NA	NA
Trimethoprim-sulfamethoxazole	0.25 (S)	NA	NA

^aThe interpretation criteria used to categorize MIC results as susceptible (S), resistant (R), and wild type (WT) are based on current breakpoints provided by the Clinical and Laboratory Standards Institute (CLSI) or MIC distribution data collected by the National Antimicrobial Resistance Monitoring System. Experiments were performed in triplicate. NA, not assessed.

A NCBI Pathogen Detection Browser query (https://www.ncbi.nlm.nih.gov/pathogens) yielded 24 qepA8-positive Escherichia coli/Shigella sequences. Most of these qepA8-positive sequences (n = 15) were of clinical or environmental origin and collected in Myanmar during August 2015 to January 2017 (8, 9). One submission had a 149,768-bp plasmid, here referred to as pM216_mF (GenBank accession no. LC492469), which carried qepA8 in a complex integron/transposon that appeared to be highly related to the context observed in 2016AM-0877 (Fig. 1). The similarities suggest that a pM216_mF-like plasmid was a potential source of the IS1-mediated composite transposon observed in the chromosome of 2016AM-0877.

FIG 1.

Comparative analysis of the genetic context of qepA8 gene in the chromosome of 2016AM-0877 (GenBank accession no. CP033510.1, reported herein) and on pM216_mF (9). Gray indicates highly related regions (<10 single-nucleotide polymorphisms in total), and striped regions indicate differing cassette arrays, where variability may be expected. Gene features are shown as arrows pointed in the orientation of the features [catA1, qepA8, sul1, mphR(A), mph(A), and mrx]. White block arrows represent insertion sequences (IS1R, IS1A, ISCR3, IS6100, and IS26), orange boxes represent conserved segments (CS) of integrons (5′-CS and 3′-CS), light-blue boxes represent gene cassettes (bla_OXA-1, aadA1, and dfrB4), and the “lollipops” represent direct repeats flanking each end of the inserted region. Unit transposons (Tn21, Tnchr, and Tn402) are shown as colored boxes, and their inverted repeat regions are shown as flags, with the flat side as the outer boundary of the transposon. Truncations are represented by a jagged edge on the truncated side(s) of the feature, and the number of bases present in between features is shown beneath the red dashed line. The diagram was generated using Galileo AMR (12) (http://galileoamr.arcbio.com/mara/).

A transformation experiment was used to assess functionality of qepA8 carried on pM216_mF, as the activity against quinolones mediated solely by qepA8 was unclear due to QRDR mutations (S→L at gyrA83, D→N at gyrA87, and S→I at parC80) in both 2016AM-0877 and the bacterial host of pM216_mF. Briefly, plasmid DNA was extracted from pM216_mF and introduced into a TOP10 E. coli strain by electroporation. Broth microdilution for the TOP10 transformants compared with untransformed E. coli TOP10 (performed in triplicate) found an at least 8-fold and 2-fold increase in the MICs for ciprofloxacin and levofloxacin (Table 1), respectively; however, MIC values remained significantly below clinical breakpoints (10).

Although the ciprofloxacin MIC did not increase above the resistance breakpoint, PMQR genes may complement other quinolone resistance mechanisms to reach clinical resistance levels and may additionally facilitate the selection of higher-level resistance by increasing the mutant prevention concentration (11). To our knowledge, this is the first report on the genetic context and functionality of qepA8. As public health laboratories transition into this era of whole-genome sequencing, there will be increasing opportunities to investigate novel variants of resistance genes and determine the extent to which they confer antimicrobial resistance.